2023 Scouting Advanced Metrics

With the 2023 season over, I’m curious if teams would be willing to share some of the more advanced metrics they collected or calculated outside of the ‘standard data’ (auto charge station, auto/tele game pieces scored, end game…).

On 1706 we did the following:
– Timed how long it took teams to enter, collect a game piece and exit the loading zone. Leveraging experience from 2017, we felt this would be critical in sorting teams that score similar number of game pieces per match as a team that is slow in the loading zone had the potential to slow down their partners as well.

– Timed how long it took teams to enter, score and leave the community. Similar to the loading zone, but less important as we preferred the ‘stay in your lane’ strategy and assigning grids to avoid congestion at the grid.

– Most experienced scouts rated robot’s driver ability on a 1-5 scale. More important earlier in the year when third robots would be playing defense, but still something we noted if a team has really high or really low at the championship.

– Calculated averaged number of game pieces scored in auto when starting on the side. This was a pretty significant factor in 1st pickability.

– Calculated percent of times a robot engaged the charge station in auto when they started in the middle position and attempted it. This was very significant in 2nd and 3rd pickability.


I absolutely love this metric for this game. Can you give more detail about how you collected it? Did each scout have a stopwatch (was this functionality built into an app solution you use)? Was only one scout responsible for each robot each match? I could imagine stopwatching each part of a cycle could get intensive for scouts over the long haul, especially if you keep track of lots of other data.

We have at least one metric worth sharing here but I’ll ping my students about that.

Yes we have an interesting formula called “Scoring Efficiency” which attempted to calculate how intelligently they placed game pieces (how well they scored links). To do this for an individual alliance the formula is:

alliance pieces scored rounded down to the nearest multiple of 3 / the number of links

Then we just averaged these for an individual team. We didn’t end up using it much but it was still a cool stat to have.


My student lead is crafting up a post about our scouting system for this year, but we had an additional 2 scouts, each scouting an entire alliance. One of their responsibilities was to collect this loading and scoring time using stopwatch functionality built into our android app. They didn’t get every cycle for every robot but they got enough to get some idea. Ideally, we would have enough students to have 1 student assigned to each robot to collect this info, but we did not want to burn our students out.


This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.